I previously talked about writing JSON exporters in Python for Prometheus. Since then, the API for custom collectors in the Python client has been improved. So let's see how easy it is to create a new exporter!
Reliable Insights
A blog on monitoring, scale and operational Sanity
Getting started with a new library it helps to have an example to work from. Let's look at a simple example of using Prometheus instrumentation in Java.
High CPU load is a common cause of issues. Let's look at how to dig into it with Prometheus and the Node exporter.
As your Prometheus usage grows and starts to get loaded, it'd be useful to know which metrics are using the most resources so that you can re-evaluate their utility.
A common question from new users is if they need to restart Prometheus every time they change the configuration. The good news is that you don't, allowing your monitoring to continue uninterrupted as your system changes.
Prometheus doesn't try to lock you into it's ecosystem - in fact it makes it straightforward to both get data both in and out. This reduces operational overhead and allows for smoother transitions between monitoring systems.
Having to manually update a list of machines in a configuration file gets annoying after a while. One of the features of Prometheus is service discovery, allowing you to automatically discover and monitor your EC2 instances!
The node exporter includes many metrics out of the box, it can't possibly cover all use cases though. That's where the textfile collector comes in, allowing you to extend machine instrumentation for your use case.
It's easy to get a demo Prometheus up and running with Docker. How do you take it beyond that and provide your own configuration?
Many of the companies I talk to either want to move off Graphite, or are already doing so. Let's look at how to get your existing data that's going to Graphite into Prometheus instead.
